_geometry3d module

class pyrockfall._geometry3d.Geometry3D(nodes: NDArray[floating], *, elements: NDArray[integer] | None = None, nodes_std: NDArray[floating] | None = None, attributes: Sequence[int] | NDArray[integer] | None = None, neighbours: NDArray[integer] | None = None)[source]

Bases: Geometry

3D triangular geometry with 3D intersection helpers (exit time & parabola hit).

This subclass keeps Geometry’s nodes/sampling logic and only overrides:

Parameters:
  • nodes – Node coordinates of shape (N, 3).

  • elements – Triangle connectivity of shape (E, 3) (0-based).

  • materialIDs – Optional per-element IDs into materials (when materials is a table).

  • nodes_std – Optional node std-devs, shape (N, 3).

  • neighbours – Optional directed-edge neighbours array of shape (E, 3). If omitted, it will be computed lazily on first access.

Raises:

ValueError – If nodes are not 3D, elements are not triangles, or provided neighbours has an invalid shape/index range.

property neighbours: ndarray

Directed-edge neighbours per triangle.

Convention: for triangle [v0, v1, v2], neighbours[e, 0] is across the directed edge v0→v1, neighbours[e, 1] across v1→v2, and neighbours[e, 2] across v2→v0. Boundary → -1.

Type:

np.ndarray of shape (E, 3)

exitTime(p: ndarray, v: ndarray, elem_id: ndarray, *, samples: ndarray | None = None, tol: float = 1e-12) → Tuple[ndarray, ndarray][source]

Earliest t>0 where x(t)=p+v t hits a directed edge of the current triangle.

Parameters:
  • p – Positions (3, S).

  • v – Velocities (3, S).

  • elem_id – Current triangle per sample (S,).

  • samples – Optional sampled nodes (3, M, 1) or (3, M, S).

  • tol – Numerical tolerance.

Returns:

next_elem (S,), t_exit (S,) with -1/nan for no exit.

Return type:

Tuple[np.ndarray, np.ndarray]

intersectParabola(p: ndarray, v: ndarray, a: ndarray, current: ndarray, *, samples: ndarray | None = None, t_min: float = 0.0, t_max: float = inf, tol: float = 1e-12, max_steps: int = 32) → Tuple[ndarray, ndarray][source]

First impact for x(t)=p + v t + 0.5 a t² using directed-edge neighbour walking.

Parameters:
  • p – Initial positions (3, S).

  • v – Initial velocities (3, S).

  • a – Constant accelerations (3, S).

  • current – Starting triangle per sample (S,) (use -1 for “unknown”).

  • samples – Optional sampled nodes (3, M, 1) or (3, M, S).

  • t_min – Minimum accepted time (inclusive).

  • t_max – Maximum accepted time (inclusive).

  • tol – Numerical tolerance.

  • max_steps – Maximum neighbour-walk iterations.

Returns:

(hit_elem (S,), t_hit (S,))

with -1/nan on no impact.

Return type:

Tuple[np.ndarray, np.ndarray]